chatgpt-apps

Scaffold ChatGPT Apps SDK projects with MCP server and widget UI.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/allenbenj/pages --skill chatgpt-apps-allenbenj
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: chatgpt-apps
Source: https://github.com/allenbenj/pages/tree/main/.agents/skills/chatgpt-apps
Command: npx skills add https://github.com/allenbenj/pages --skill chatgpt-apps-allenbenj

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @modelcontextprotocol/ext-apps, @modelcontextprotocol/sdk, zod, and includes scripts (resource) components.

What problem does it solve?

Build, scaffold, refactor, and troubleshoot ChatGPT Apps SDK applications that combine an MCP server and widget UI. Use when Codex needs to design tools, register UI resources, wire the MCP Apps bridge or ChatGPT compatibility APIs, apply Apps SDK metadata or CSP or domain settings, or produce a docs-aligned project scaffold. Prefer a docs-first workflow by invoking the openai-docs skill or OpenAI developer docs MCP tools before generating code.

Core Features & Use Cases

  • Classify app requests into a small set of archetypes and identify the closest upstream starting point (official example, ext-apps starter, or local fallback).
  • Plan tools, architecture, and repository layout for server and widget code, including tool definitions, CSP/URI versioning, and metadata alignment.
  • Generate or patch a minimal MCP server + vanilla widget scaffold with local dev and tunnel setup guidance, plus validation against the minimum working repo contract.
  • Provide doc-references and workflow guidance to ensure docs-first parity and submission-readiness where applicable.

Quick Start

Generate a starter scaffold with the provided generator, then tailor tool names, descriptions, CSP, and URI versioning to your app concept.

Frequently Asked Questions about chatgpt-apps

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I scaffold a ChatGPT Apps SDK project with an MCP server and widget UI?

Generate a starter scaffold using the provided generator, then tailor tool names, descriptions, CSP, and URI versioning to your app concept while validating against the minimum working repo contract.

What is the best way to plan tools and architecture for a ChatGPT Apps SDK application?

Apply a docs-first workflow by invoking the openai-docs skill or OpenAI developer docs MCP tools before generating code to ensure your tool planning and architecture design align with the latest contracts.

Can I use the ext-apps starter to validate my existing ChatGPT Apps SDK repository?

The validation process checks your repository against the minimum working repo contract, ensuring server and widget code, CSP settings, URI versioning, and metadata all comply with the latest ChatGPT Apps standards.

Do I need to follow a docs-first workflow when building ChatGPT Apps SDK applications?

You should follow a docs-first workflow by fetching the latest OpenAI developer docs before generating code, ensuring your scaffolded ChatGPT Apps SDK application maintains strict parity with official contracts and metadata requirements.

How does CSP and URI versioning work when scaffolding a ChatGPT Apps SDK?

CSP and URI versioning are configured during scaffolding to secure the widget UI and manage endpoint compatibility, ensuring the generated MCP server and widget adhere to the latest ChatGPT Apps SDK contracts.

What are the limitations of adapting upstream examples for ChatGPT Apps SDK projects?

Adapting upstream examples risks contract drift if the starting point diverges from the latest official docs, requiring you to patch tool definitions, CSP settings, and metadata to maintain ChatGPT Apps SDK repo compliance.